[ILUG] dying hard disc - upgrade questions?

[Paul Jakma]

On Mon, 17 May 2004, Brian Foster wrote:

>    the issue here is that, AFAIK, ATA-1/-2 have a limit
>   of c.137Gb, suggesting my replacement must be ≤ 120Gb,

Linux supports LBA48, so you can get whatever size disk you want. 
Though, I'm not sure what happens with respect to BIOSen.

I have a 160GB disk in an ancient Deskpro at the moment, and linux 
has no problems with it, but I have not booted from it.

>   QUESTION:  how can I work out if my controller et.al.
>             is ATA-1, ATA-2, ATAPI, or something else?

Drives generally do not use ATAPI. The biggest issue will be BIOS.
 
>   problems (e.g., on power failure), albeit if the
>   driver/ATA/IDE/cache(/BIOS?) supports flush and barrier
>   operations, then it is perhaps Ok.

The problem is drives which lie about write-caching (write-back, 
write-through doesnt matter) and default enable it and/or lie when 
asked to disable it. Even worse are drives which lie about the flush 
command.

This problem is a time-bomb no matter what the OS, so drive
manufacturers have copped on, recent/new drives should not lie about
write-cache.
 
>   QUESTION:  should I steer clear of the drives with
>             “cache‟, or are they generally Ok?

Cache is always good :) (lying about write caching excepted). It's
also just about required for large drives.
 
>   assume is the master), as the slave(?) --- but then can
>   I boot _exclusively_ from the slave without the master
>   being involved (or even present?)?

That depends on the BIOS. If master is not present, slave will be
0x80 (or is first disk 0x81? cant be bothered to read lilo docs) and
BIOS will try boot from it. Handiest thing to do is to install grub
on both boot sectors. As long as you can get to a grub prompt you can
always boot.
 
regards,
-- 
Paul Jakma	paul at clubi.ie	paul at jakma.org	Key ID: 64A2FF6A
	warning: do not ever send email to spam at dishone.st
Fortune:
Editing is a rewording activity.